Bge M3 Unsupervised
Jan 2024
Bge M3 UnsupervisedCompare inon
Bge M3 Unsupervised measured on 10 of 261 Embeddings benchmarks10
BenchmarkByScoreRank
Reranking5
Other5
No benchmark matches
price not verifiedprice
44.1RTEB French beta
Jan 2024released
1,024Dimensions
8K tokensMax input
Alternatives?
ModelScoreCostPriceWhat changes
BetterQwen 3 Embedding 8BunstatedOpen+35%—$0.01Memory (4-bit): 0.3 GB → 4.5 GB
OpenNemotron 3 Embed 8B BF16+47%——Memory (4-bit): 0.3 GB → 4.8 GBBetterBge M3unstatedOpen+14%—$0.01iBetterMultilingual E5 LargeunstatedOpen+7%—$0.01Max input: 8K tokens → 1K tokensBetterNemotron 3 Embed 1B BF16Open+29%—$0.015Memory (4-bit): 0.3 GB → 0.7 GBModelScoreCostPriceWhat changes
Nothing here yet.
ModelScoreCostPriceWhat changes
Qwen 3 Embedding 8BunstatedOpen+35%—$0.01Memory (4-bit): 0.3 GB → 4.5 GBBge M3unstatedOpen+14%—$0.01iMultilingual E5 LargeunstatedOpen+7%—$0.01Max input: 8K tokens → 1K tokensNemotron 3 Embed 1B BF16Open+29%—$0.015Memory (4-bit): 0.3 GB → 0.7 GB
Jina Embeddings v5 Text NanounstatedOpen+14%—$0.02ModelScoreCostPriceWhat changes
Nemotron 3 Embed 8B BF16Open+47%——Memory (4-bit): 0.3 GB → 4.8 GBOOcten Embedding 4BOpen+47%——Memory (4-bit): 0.3 GB → 2.4 GBOOcten Embedding 4B INT8Open+46%——Memory (4-bit): 0.3 GB → 2.4 GBNV Embed v2unstatedOpen+39%——Memory (4-bit): 0.3 GB → 4.7 GBOOcten Embedding 8BOpen+39%——Memory (4-bit): 0.3 GB → 4.5 GBModelScoreCostPriceWhat changes
Bge M3unstatedOpen+14%—$0.01Bge Large EnunstatedOpen-37%——Max input: 8K tokens → 1K tokensBge Large En v1.5unstatedOpen-40%—$0.01Max input: 8K tokens → 1K tokensBge Base En v1.5unstatedOpen-47%—$0.005Max input: 8K tokens → 1K tokensBge Base EnunstatedOpen-53%——Max input: 8K tokens → 1K tokensBge Small En v1.5unstatedOpen-61%—$0.02Max input: 8K tokens → 1K tokensBge Base Zh v1.5Open-85%——Max input: 8K tokens → 1K tokensBge Large Zh v1.5Open-88%——Max input: 8K tokens → 1K tokensAll BAAI models →Run it yourself
estimateFits on✓ 8 GB (phone, laptop) · full quality✓ 16 GB · full quality✓ 24 GB GPU · full quality✓ 48 GB GPU · full quality✓ 80 GB GPU · full quality
LicenceMITfree for commercial use
Benchmarks
in Embeddings & RetrievalFAQ
How much does Bge M3 Unsupervised cost?›
There is no per-unit API price: price not verified.
How good is Bge M3 Unsupervised, and what is it best at?›
#54 of 212 in Embeddings.
Is Bge M3 Unsupervised better than Text Embedding 3 Small?›
On RTEB French beta · Mean over tasks · multilingual (2 languages), Bge M3 Unsupervised scores 44.1 vs 44.2 — about the same.
When was Bge M3 Unsupervised released? Is there a newer version?›
Released 2024-01-30. It is the current version in its line.
Is Bge M3 Unsupervised open source?›
Yes, the weights are open (MIT licence).
Can I run Bge M3 Unsupervised locally?›
The weights are open and can be self-hosted; hardware needs are not confirmed yet.
What is Bge M3 Unsupervised?›
Bge M3 Unsupervised is an open-weights embeddings & retrieval model from BAAI, released Jan 2024.
What are the best alternatives to Bge M3 Unsupervised?›
Qwen 3 Embedding 8B (better), Nemotron 3 Embed 8B BF16 (open weights).